PROJECT ABSTRACT The goal of this proposal is to test the fidelity of three types of patient-derived breast cancer models with regard to genomic and epigenetic aberrations, clonal heterogeneity and evolution, and treatment response/resistance. We will compare PDXs, patient-derived organoids (PDOs), and patient-derived conditionally-reprogrammed cells (PDCRCs) from a total of 22 patients with breast cancer. The models will be derived from patient samples acquired at 2-3 longitudinal time points during the patients? cancer treatments. To our knowledge, this will be the first effort to establish a triad of patient-derived models of cancer (PDMCs) in longitudinal series from breast cancer patients undergoing standard clinical care. Importantly, the models will be associated with annotated clinical data on patient treatment and outcomes. PDMCs will be functionally evaluated for their response to patient-matched therapies. Aim 1 is focused on testing whether PDMCs from patients undergoing therapy replicate clinicopathological, molecular, genomic and cellular phenotypes observed in the patients? clinical samples. PDMCs will be generated from viable breast cancer specimens obtained prior to and following patient therapy ? either in the neoadjuvant or metastatic setting. We will generate a triad of patient-matched PDMCs (PDX, PDO, PDCRC) and compare tumor pathology, gene expression, WGS/WES, DNA methylation, CNV, mutation profiles, and cellular clonality/heterogeneity between the patient tissue and PDMCs. Aim 2 will investigate whether PDMCs appropriately model patient response to therapy ? an assessment that is needed to determine their potential application in basic research, drug discovery, and as predictors of optimal therapies for patients undergoing treatment. PDMCs will be evaluated in a co-clinical study to test the concordance of response between PDMCs and the clinical response observed in patients. We will also evaluate the concordance of PDOs and PDMCs against a panel of FDA-approved cancer therapies, genomic-guided therapies, and investigate whether the chemo-sensitivity of clones in PDOs and PDCRCs is associated with their observed clonal frequency in patient tumors following treatment. Our study will not only determine whether PDMCs can model the repertoire of breast cancer phenotypes, but will also determine, through a co-clinical study, whether they can functionally replicate patient response to therapies.